WO1995027235A1 - Systeme et methode de regulation - Google Patents

Systeme et methode de regulation Download PDF

Info

Publication number
WO1995027235A1
WO1995027235A1 PCT/JP1995/000612 JP9500612W WO9527235A1 WO 1995027235 A1 WO1995027235 A1 WO 1995027235A1 JP 9500612 W JP9500612 W JP 9500612W WO 9527235 A1 WO9527235 A1 WO 9527235A1
Authority
WO
WIPO (PCT)
Prior art keywords
calculated
normalized
value
control
deviation
Prior art date
Application number
PCT/JP1995/000612
Other languages
English (en)
Japanese (ja)
Inventor
Mamoru Egi
Katuyuki Inage
Original Assignee
Omron Corporation
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Omron Corporation filed Critical Omron Corporation
Priority to EP95913388A priority Critical patent/EP0753802A4/fr
Priority to KR1019960705531A priority patent/KR970702515A/ko
Publication of WO1995027235A1 publication Critical patent/WO1995027235A1/fr

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/0265—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion
    • G05B13/0275—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion using fuzzy logic only
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B11/00—Automatic controllers
    • G05B11/01—Automatic controllers electric
    • G05B11/32—Automatic controllers electric with inputs from more than one sensing element; with outputs to more than one correcting element
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D23/00—Control of temperature
    • G05D23/19—Control of temperature characterised by the use of electric means
    • G05D23/1917—Control of temperature characterised by the use of electric means using digital means

Definitions

  • control target In a system that controls the system, the control target is stable from one state (for example, the initial state) to another state (for example, the target state).
  • This control is roughly divided into the following two items concerning the control system and the method of controlling the transition.
  • One of them is to change the control target to the target state stably by modifying the target value given to the controller that indicates the target state. It is something to be done.
  • the other one should be given to a controller D-controller that represents the transition path in order to control the control target to transition from the initial state to the target state.
  • the controlled object can be stably transitioned to the target state.
  • a temperature control system As an example of a control system that modifies a target value that should be given to a controller that indicates a target state, a temperature control system is used. There is a time.
  • the heating furnace, heater plate, etc. are divided into multiple areas (a typical point in this area is called a heating point).
  • the heat is set for each area, and the temperature control is independently performed for each area so that each heating point has the desired target temperature.
  • the target temperature is set for each area according to the product to be heated, the product to be molded, and so on.
  • the temperature of each area may be affected by the heat (interference) generated from the adjacent area, and an overshoot may occur.
  • an overshoot may occur in a tunnel type heating furnace.
  • a steady-state deviation remains, and the temperature in the heating furnace returns to the target temperature. It takes time to complete. In other words, the settling time becomes longer.
  • Temperature control device to prevent such interference Special Publication No. 59-35 212 Publication and Special Publication No. 61-45606 Publication No. 61-45606 and Japanese Patent Publication No. 61-155 It is described in JP-B-6-316. These temperature control devices model the interference between the heating points and use this model. The purpose of this is to determine the amount of operation using a device and to prevent a thousand square meters.
  • the temperature at each heating point is measured in order to model the interference between the heating points. Using this measured temperature, a parameter of the transfer function is calculated. As the heating point increases, it takes more time to calculate the transfer function that represents the model.
  • a sub-target value representing the transition path is given to the controller.
  • An example of such a control system is a position control system in a crane or the like.
  • the specified speed pattern is a pattern in which the load is accelerated, decelerated, and decelerated so that the load swing becomes as small as possible.
  • This invention is based on the assumption that the control target is in a certain state (for example, in the initial state).
  • the purpose is to control so as to make a stable transition from) to another state (for example, the target state).
  • control target is modified by modifying the target value given to the controller representing the final target state.
  • the actuator is an electromagnet, a magnetic sensor using a ball element is used as a sensor, and the spatial distribution of the magnetic flux density is a predetermined magnetic flux ratio. It is controlled to be a turn.
  • an imaging element such as a CCD camera or a light receiving element is used as a sensor when the evening is a light emitter, and the spatial distribution of illuminance is limited. It is controlled so that it has a constant illumination level.
  • the control system according to the first invention is provided so as to correspond to each of a plurality of sky nights existing spatially and the above plurality of nights.
  • the temperature measured by the device and the plurality of temperature sensors was generated by the target temperature generator described above.
  • the manipulated variables are calculated to approach the corrected target temperature, and the manipulated variables are given to the above-mentioned plurality of heaters, respectively.
  • the target temperatures are measured by the above-mentioned plurality of temperature sensors, and the measured temperature is determined by the corresponding final target temperature.
  • a normalized normalized measured temperature is calculated, a reference value is determined based on the normalized measured temperature, and the normalized measured temperature is compared with the reference value.
  • the reference value deviation calculated by the reference value deviation calculating means for calculating the reference value deviation and the reference value deviation calculated by the above-described reference value deviation calculating means, respectively, is differentiated in time.
  • the differential differential value calculation means for calculating the differential differential value by using the reference differential deviation calculation method All the normal values are calculated based on the reference value deviation calculated by the means and the deviation differential values calculated by the above-mentioned differential differential value calculating means.
  • the compensation gain is calculated by the compensation gain calculation means and the temperature sensors described above.
  • the deviation between the measured temperature and the corresponding final target temperature is normalized by the above-mentioned final target temperature, and the normalized deviation is calculated.
  • Means, the correction gain calculated by the correction gain calculating means, and the settling degree calculation And San out has been settling degree Tsu by the stage, and have based Dzu to a predetermined stand above gain Lee down your good beauty Jo Tokoro settling gain Lee down, compensation
  • the above-mentioned final target temperature is obtained by using the correction amount calculating means for calculating the respective amounts, and the correction amount calculated by the correction amount calculating means.
  • the target temperature calculation means for calculating the above-mentioned target temperature by correcting the temperature is provided.o
  • the control method according to the first invention is provided so as to correspond to each of the plurality of heaters existing in space and the plurality of heaters described above.
  • a plurality of temperature sensors for measuring the temperature of the controlled object, and the measured temperatures measured by the plurality of temperature sensors described above, respectively, are provided.
  • the controller calculates the manipulated variables so as to approach the target temperature, and provides a controller that gives these manipulated variables to the heaters.
  • the measured temperatures measured by the above-mentioned temperature sensors are normalized by the corresponding final target temperatures, respectively. Is calculated, and a reference value is determined based on these normalized measurement temperatures, and each normalized measurement temperature is calculated.
  • the reference value deviation from the above reference value is calculated respectively, and the calculated reference value deviation is differentiated with respect to time to calculate the differential deviation value.
  • the corrected target temperature can be calculated.
  • the measured temperatures measured by a plurality of temperature sensors are each diverted by the corresponding final goal / or endowment. It is normalized and the normalized measurement temperature is calculated.
  • the normalized measured temperature indicates how much the temperature of the control target has reached the target (final target temperature). That is, the normalized measurement temperature indicates the degree of response. This indicates that the closer the normalized measured temperature is to 1, the closer the temperature of the controlled object is to the target temperature. Since the measured temperature is normalized by the final target temperature, the final target temperature is not spatially uniform and each of them has a different value. For such a notch, the spatial distribution of temperature is random and the work force ⁇ Suitable, with a certain gradient, stepped, etc.
  • a reference value is determined based on these normalized measured temperatures.
  • the reference value is, for example, the average value of all normalized measured temperatures in the transient state, and in the steady state. Is 1 * * 0
  • the transient state is when all the normalized measured temperatures are 1 or less, and the normal state is any one of the normal states. • When the measured temperature exceeds 1 •.
  • a reference value deviation between each normalized measurement temperature and the reference value is calculated, and the reference value deviation is, for example, the case where the reference value is the average of the normalized measurement temperature. This indicates whether the response of the temperature at the position where the controlled object is responding in an average manner to the temperature at the other position is faster or slower.
  • the reference value deviation is time-differentiated in each case, and the differential deviation value is calculated in each case.
  • the differential deviation value indicates whether the reference value deviation tends to decrease or increase.)
  • the temperature response at one of the locations will tend to be faster or slower or the same as the temperature response at other locations. It shows whether or not.
  • a correction gain is calculated and calculated so that all the normalized measured temperatures have the same value. It is. For example, if the reference value deviation is a large positive value, the correction gain is set to a large negative value. In addition, if the reference deviation is a large positive value and the reference deviation is a large positive value, the correction gain becomes a large value. It is set. On the other hand, when the reference value deviation and the 1 ⁇ differential value are both almost equal to 0, the correction gain is also almost equal to 0, and the correction value is also equal to 0. The deviation between the measured temperature measured by the plurality of temperature sensors and the corresponding final target temperature is normalized by the final target temperature. , And the normalized deviation is calculated respectively.
  • the settling degree is calculated based on the normalized deviation.
  • the degree of settling represents the force at which each measured temperature is at start-up or at set-point. The closer the normalized deviation is to 1, it is at set time, and the farther it is from 1, the more it is at startup.
  • the rise gain is a gain for correcting the correction gain at the time of startup
  • the setting gain is a correction gain for correcting the correction gain at the time of settling. It is a gain for simplicity.
  • a rise gain or a stabilization gain is used, and the gain is corrected by these gains, and the correction amount is calculated. Will be issued.
  • the corrected target temperature is calculated by correcting the final target temperature by using the correction weight. If the final temperature deviation is positive and large, the corrected target temperature becomes the final target. If the reference temperature is set lower than the target temperature and the reference him difference is large negatively, the corrected target temperature will be set higher than the final target temperature. For those whose quasi-value deviation is positive and large (the response speed is fast), the response target speed is slowed down by setting the correction target temperature small, and the reference value is reduced. If the deviation is positive and large (the response speed is For the slower ones, the response speed is set to be higher by setting the target temperature higher. In other words, the proportion of the response of the measured temperature to the corresponding final target temperature is the same in the control space.
  • the corrected target temperature is generated so that the response speed of the temperature at each position becomes the same value, so that the influence from the adjacent light source is obtained. Interference) Even if received, the rate of change in temperature will be uniform. This makes it harder for one-shot to occur due to mutual interference. In particular, in the case of a tunnel type furnace, the settling time is shortened by controlling such that overshooting does not occur. Can be obtained.
  • the control is performed for each control target such as a tunnel-type force, a heating furnace, a heater, and a heater.
  • the need for modeling is eliminated, and the same control can be applied to various control targets.
  • the control system which is a generalization of the first invention, comprises a plurality of spatially existing actuaries for driving a controlled object.
  • a plurality of sensors and a plurality of sensors are provided corresponding to each of the factories to measure the control amount of the above-mentioned control target. Measured by the sensor
  • Each of the controlled variables is calculated, and the manipulated variables are calculated so as to approach the given target value.
  • These manipulated variables are calculated by the plurality of actuators described above.
  • the control amount measured by the above-mentioned plurality of sensors is used.
  • Introductory note D is characterized by the fact that it has a target value generation device
  • the control method that generalizes the first invention is that a plurality of spatially existing actuators for driving a controlled object are provided. It is set up corresponding to each of the evenings, and is directly connected to the control target of the above-mentioned controlled object: a plurality of sensors to be measured, and a plurality of sensors to be measured.
  • the control variables measured by the controller are calculated, and the control variables are calculated so as to approach the target value, and the control variables are described above.
  • a controller is provided for each of the plurality of factories overnight, and the control amount measured by the above-mentioned plurality of sensors is provided at the end of the application. Calculate the normalized control values normalized by the standard values, and calculate the reference values by controlling these normalized controls.
  • the reference amount deviation between each normalized control amount and the above reference amount is calculated, and based on each reference amount deviation, all the normalized control amounts have the same value.
  • the above-mentioned target values are calculated by modifying the above-mentioned final target values so as to obtain the target values, and these target values are converted to the above-mentioned control values. It is given to the ruler.
  • the actuating element is a heater
  • the sensor is a temperature sensor
  • the control variables measured by a plurality of sensors are each normalized by the final target value of the corresponding endowment. Then, the normalized control amount is calculated.
  • the normalized control amount indicates the extent to which the control amount has reached the target. In other words, the normalized control amount represents the degree of response, and the closer the normalized control amount is to one, the more the target is reached. Since the control amount is normalized by the final target value, a setting is made so that the final target values are not uniform and each of them has a different value. can do.
  • Such patterns include those with a random spatial distribution of the final target values, those with a certain gradient, and those with a stepped shape. , Etc.
  • the reference amount is determined based on these normalized control amounts.
  • the standard quantity is the average value of all normalized control quantities in the transient state, and is 1 in the steady state.
  • a transient state is defined as a state in which all the normalized control s force s 1 or less. When one of the normalized controlled variables exceeds one, the steady state is exceeded.
  • the reference amount deviation between each normalized control amount and the reference amount is calculated respectively, and the reference amount deviation is, if the reference amount is equal to the average value of the normalized control amount,
  • the table shows whether the response of the control amount at the position where the control object responds to the average response is faster or slower than the control amount at the other position. are doing.
  • the responses of the control quantities are adjusted so that all the normalized control quantities have the same value, that is, the respective control quantities. Is modified so that the percentage approaching the corresponding final target is the same in the control space. Is calculated.
  • the target value is set to be smaller than the final target value when the reference value deviation is positively large, and the reference value deviation is negative, and when the reference value deviation is large, the final value is small.
  • the response speed is slower for the control amount with a faster response speed by setting the target value smaller.
  • the response speed is increased by setting a large target value for the control amount with a slow response speed.
  • the target value is generated so that the response speed becomes the same value for each control amount, so that the adjacent actuators are set. Even if the influence (interference) is received, the rate of change in the control amount will be uniform. As a result, the control amount overshoot or undershoot due to the interference can be achieved. The shot is less likely to occur. Therefore, the settling time is shortened by preventing the occurrence of one shot or one shot. can do.
  • the target value generation device corresponds to the control amount measured by the plurality of sensors described above, and corresponds to each of the control amounts measured by the target sensors.
  • the above-mentioned normalized control amounts are calculated, and the above-mentioned reference amounts are determined based on these normalized control amounts.
  • the above target values are calculated and calculated. This is realized by means of calculating the standard value.
  • the target value generating device includes a reference value deviation calculated by the reference value deviation calculating means.
  • a differential differential value calculation means for calculating the differential differential value by performing the intermediate differentiation is further provided, and the target value calculating means is provided in the above-described manner.
  • the reference amount deviation calculated by the reference amount deviation calculating means and the differential differential value calculated by the 1 ⁇ 21 difference differential value calculating means described above are calculated based on the reference amount deviation calculated by the reference amount deviation calculating means.
  • the above target values are calculated by modifying the final target values so that all the normalized control values have the same value. It is something.
  • each reference quantity deviation is further subdivided in time, so that the differential deviation value is further calculated.
  • the deviation derivative value indicates whether the deviation of the normalized control variable with respect to the reference variable tends to decrease or to increase. In other words, the response speed at that position tends to be faster or slower, or the same as the response speed at other positions. It shows whether there is any stiffness.
  • the target value is calculated using the differential value of the deviation. If the deviation derivative is positive, the target value is set smaller, and if negative, the target value is set larger.
  • the target value calculated in this way is given to the controller.
  • the differential value of the deviation is considered in addition to the deviation of the reference value, so that the target value can be calculated more accurately.
  • the target value generation device includes a reference amount deviation calculated by the reference amount deviation calculating means.
  • a correction gain calculation for calculating a correction gain based on the difference and the deviation differential value calculated by the above-described differential differential value calculation means, respectively.
  • Outgoing means multiple above
  • the deviation between the control amount measured by the sensor of the above and the corresponding final target value is normalized by the above-mentioned final target value.
  • the normalized deviation calculating means for calculating the normalized deviation, and the normalized deviation calculated by the normalized deviation calculating means are based on the normalized deviation calculated by the normalized deviation calculating means.
  • a settling degree calculating means for calculating each of the settling degrees is further provided, and the above-mentioned target value calculating means is provided by the above-mentioned corrected gain calculating means. Based on the calculated correction gain and the settling degree calculated by the settling degree calculating means described above, all the normalized control amounts are the same.
  • the above-mentioned target values are calculated by modifying the above-mentioned final target values so as to obtain the same values.
  • the correction gain is set so that all the normalized control amounts are the same based on each reference value deviation and the deviation differential value. Is calculated accordingly. For example, if the reference value deviation is a large positive value, the correction gain is set to a large negative value. If the reference value deviation is a large positive value and the reference value deviation is a large positive value, the correction gain is set to a large value. It is. On the other hand, when the reference value deviation and the deviation differential value are almost equal to zero, the correction gain is also almost equal to zero.
  • the deviation between the control value measured by the multiple sensors and the corresponding final target value is the normal deviation depending on the final target value. Then, the normalized deviation is calculated. So The settling degree is calculated based on the normalized deviation. The settling degree indicates whether each control amount is at the time of startup (temperature state) or at the time of settling (steady state). The closer to 1 the normalized deviation is at set time, the farther it is from 1 it is at startup.
  • the correction gain is modified according to the value of the settling degree, and the correction gain is adjusted based on the corrected correction gain.
  • the target value is calculated.
  • the target value calculated in this way is given to the controller.
  • the target value is calculated in consideration of the degree of settling, and the transient state is calculated.
  • an appropriate target value can be calculated according to a steady state.
  • the target value generating device includes the correction gain calculated by the correction gain calculating means.
  • the amount of correction is determined based on the gain, the settling degree calculated by the settling degree calculating means, and a predetermined start-up gain and a predetermined setting gain.
  • a correction amount calculating means for calculating the correction value is further provided, and the target value calculating means described above is provided with the correction amount calculated by the correction amount calculating means. By modifying the final target value using a positive quantity, the target value can be calculated.
  • the correction gain, the settling degree, and the start-up The correction amount is calculated based on the gain and the settling gain.
  • the rising gain is a gain for correcting the correction gain at the time of startup
  • the setting gain is a correction gain at the time of setting. It is a gain to correct the problem.
  • a rise gain or a settling gain is used to calculate the amount of correction in which the correction gain has been corrected. .
  • the target value is calculated by correcting the final target value using the correction amount.
  • the target value calculated in this way is given to the controller.
  • the target value is set using the desired start-up gain or settling gain. Can be adjusted.
  • the second invention is based on the modification of the sub-goal value that should be given to the controller that represents the transition path from the initial state to the target state. This is to make the control target stably transition to the target state.
  • the second invention is applied to position control, attitude control, etc.
  • the factories there are motors and other driving equipment, and position sensors, speed sensors, etc. are used as sensors.
  • the control system according to the second invention is composed of a plurality of driving devices for driving a controlled object, and a plurality of the above driving devices.
  • a plurality of position sensors for measuring the position of the object to be controlled and the given final position and the specified command speed are provided correspondingly.
  • the modified command speed is generated for each driving device based on the speed and the command speed is generated by the device, and measured by the plurality of position sensors described above.
  • the manipulated variables are adjusted so that the speeds derived based on the determined positions are the corrected command speeds generated by the command speed generator described above, respectively.
  • the above-mentioned command speed generating device is provided with a controller for calculating each of the operation amounts and providing these operation amounts to the plurality of driving devices.
  • the initial position measured by the plurality of position sensors and the final target position measured by the plurality of position sensors Before or at the start of control, the initial position measured by the plurality of position sensors and the final target position measured by the plurality of position sensors.
  • the maximum distance calculating means for calculating the maximum distance based on the above, and after the control is started, the measurement is performed by the initial position and the plurality of position sensors. The distances to the measured measurement positions are calculated by using the corresponding maximum distances, and the normalized distances are calculated. The normalized distances are calculated based on these normalized distances.
  • the reference value is determined by the reference value deviation calculating means and the reference value deviation calculating means for calculating the reference value deviation between each normalized distance and the above-mentioned reference value, respectively.
  • the differential differential value calculating means for calculating the differential differential value by time-differentiating the reference value deviation calculated as described above, and the above-described standard value deviation calculating method
  • y de deviations differential value and based on Dzu Ni Let 's normalized distance of Te to base is that Do the same value, correction gain Lee down On the basis of the corrected gain calculating means for calculating the corrected gain and the corrected gain calculated by the corrected gain calculating means, respectively.
  • a command speed calculation means for calculating the above-mentioned modified command speed by modifying the above-mentioned command speed is provided.
  • the control method according to the second invention corresponds to each of a plurality of driving devices for driving a controlled object and each of the above-mentioned plurality of driving devices.
  • a plurality of position sensors that are installed and measure the position of the controlled object, and the positions measured by the plurality of position sensors. The operation amounts are calculated so that the speeds derived on the basis of the respective speeds will be the given correction command speeds, and these operation amounts are described above.
  • a controller is provided for each of the plurality of drive units, and before or at the start of the control, the above-mentioned plurality of position sensors are used. The maximum distance is calculated based on the initial position measured as described above and the final target position, and after the control is started, the above-mentioned initial position is calculated.
  • the maximum distance between the initial position and the target position measured by the plurality of position sensors is respectively determined. Is calculated.
  • the distance between the initial position and the measurement positions measured by the plurality of position sensors is normalized by the maximum distance between them.
  • the normalized normalized distance is calculated accordingly. Since the maximum distance that the drive unit can move is different, the distance moved from the initial position is normalized by the maximum distance. It indicates how far the target distance has reached the target position.
  • a reference value is determined based on these normalized distances.
  • the reference value is, for example, a normalized distance corresponding to the driving device having the largest maximum distance.
  • the movement distance is the longest because the movement time is long and the response speed is the shortest.
  • the reference value deviation between each normalized distance and the reference value is calculated.
  • the reference value deviation is a deviation from a reference value (a normalized distance having the largest distance is the largest), and is moving faster than the reference value. Or.
  • the differential value of the reference value is differentiated over time to Is calculated for each.
  • the deviation derivative value indicates whether the reference value deviation tends to increase or decrease.
  • Correction gains are calculated based on the reference value deviation and the deviation differential value so that all normalized distances have the same value. It is. When the reference value deviation is positively large, the correction gain is set to be large, and when the reference value deviation is also positively large, In addition, the correction gain is set to a large value.
  • the correction command speed is corrected based on the correction gain, whereby the correction command speed is calculated.
  • the specified command speed is, for example, a speed pattern for accelerating, constant speed, and decelerating.
  • the command speed calculated in this way is given to the controller.
  • the controlled object can be moved stably.
  • a control system that generally prescribes the second invention includes a plurality of actuators for driving the control object, and a plurality of actuators for driving the control object.
  • a plurality of first and second control units are installed to correspond to each of the evenings and measure the first and second control amounts of the control target, respectively.
  • a sub-target value generating device that generates a sub-target value to be given to the controller every night of the factories, and a second sub-unit of the above-mentioned plural targets
  • Each of the second control amounts measured by the sensor is close to the modified sub-target value generated by the above-described sub-target value generating device.
  • the controller is provided with a controller for calculating the manipulated variables as described above and providing the manipulated variables to each of the plurality of actuators.
  • the sub-target value generation device described above calculates the initial value measured by the plurality of first sensors and the final value.
  • the maximum change amount is calculated based on the target value, and after the control is started, the maximum change amount is calculated according to the above initial value and the plurality of first sensors.
  • the amount of change from the measured first control amount is calculated according to the corresponding maximum amount of change, and the normalized amount of change is calculated.
  • the reference amounts determined on the basis of the normalized change amounts are determined, and the reference amount deviation between each normalized change amount and the above reference amount is determined.
  • the control method which generally defines the second invention is a plurality of actuators for driving the control target, and a plurality of actuators for driving the control object. And a plurality of first and second units for measuring the first and second control amounts of the above-mentioned control target, respectively.
  • the sensor and the second control amount measured by the plurality of second sensors are each approaching the assigned sub-target value.
  • a controller is provided to calculate the operation amounts and to provide these operation amounts to the above-mentioned multiple factories. Before or at the start of control, based on the initial values measured by the plurality of first sensors and the final target value, Then, the maximum change amount is calculated respectively, and after the control is started, the initial value and the first values measured by the plurality of first sensors are calculated.
  • a reference amount determined based on the normalized amount is determined, and a standard amount deviation between each normalized change amount and the above standard amount is calculated, and each standard amount deviation is calculated.
  • the modified sub-target value is generated by modifying the sub-target value so that all the normalized change amounts have the same value based on the sub-target value. is there.
  • the first sensor and the second sensor are the same, and the first control amount is equal to the first control amount. It is calculated based on the control amount of 1.
  • the above actuator is a driving device, and the first sensor is a position sensor.
  • the first control variable is position
  • the second control variable is speed
  • the maximum change is based on the initial value and the final target value measured by the plurality of first sensors. Each is calculated.
  • the change between the initial value and the first control amount measured by the plurality of first sensors corresponds to the corresponding maximum change amount.
  • each of them is normalized, and the amount of change in normalization is calculated.
  • By normalizing the variation of the control variable from the initial value by the maximum variation it shows how much the normalized variation has reached the target. And review.
  • the reference amount is determined based on these normalized change amounts.
  • the reference amount is, for example, the normalized change with the largest change. This is because the control with the largest change amount is the slowest in the response of the control amount, and the response speed to the target becomes the minimum. . If the performance of the actuary is different, that performance is also taken into account.
  • the reference amount deviation between each normalized change amount and the reference amount is calculated, respectively.
  • the reference value deviation is the reference value (the maximum change is This shows how fast or slow it responds to a large normalized change).
  • the sub-target value is modified by modifying the sub-target value so that all the normalized change amounts are the same based on each reference value deviation.
  • the value is calculated for each. If the deviation of the reference value is positively large, the modified sub-target value is set small, and if the deviation of the reference value is negatively large, the modified sub-target value is increased. It is set.
  • the sub-target value may be a constant value or a value having a predetermined noise.
  • the modified sub-target value calculated in this way is given to the controller.
  • any number of actuaries can be performed during the period from the initial state (initial value) to the target state (final target value). Even if a disturbance occurs, the modified sub-target value is calculated so that the response speed is the same, so that the control target can be controlled stably. .
  • the sub-target value generation device is configured to perform the above-mentioned initial value measured by the plurality of first sensors and the above-mentioned final target before or at the start of control.
  • the change amount calculating means for calculating the maximum change amount based on the value and the control value are inputted to the initial value and the plurality of first sensors after the start of the control.
  • the above-described normalized change amount obtained by normalizing the change amount from the first control amount measured by the above-described method according to the corresponding maximum change amount, respectively. Calculated and based on these normalized variables,
  • the reference amount is determined, and the reference amount deviation calculating means for calculating the reference amount deviation between each normalized change amount and the reference amount described above, and the reference amount deviation calculating means described above.
  • the sub-target value is modified based on the reference deviation calculated by the means so that all the normalized variables have the same value. Accordingly, the above-described modified sub-target values are realized by the target value calculating means for calculating the respective sub-target values.
  • the sub-target value generation device includes a reference value deviation calculated by the reference value deviation calculating means.
  • a differential differential value calculating means for calculating the differential differential value by separating the differential values is further provided, and the target value calculating means is provided as described above. All based on the reference amount deviation calculated by the reference amount deviation calculating means and the differential differential value calculated by the above-described differential differential value calculating means.
  • the differential value of the reference quantity is differentiated with respect to time, whereby the differential value of the difference is calculated.
  • the differential deviation value indicates whether the reference value deviation tends to increase or to decrease.
  • the predetermined sub-target value is modified so that all the normalized changes are the same value.
  • the corrected sub-target values are calculated accordingly. It is.
  • the modified sub-target value is set so as to be smaller.
  • the sub-target generation device includes a reference amount deviation calculated by the reference amount deviation calculating means. Based on the difference and the differential differential value calculated by the differential differential value calculating means, all the normalized change amounts become the same value.
  • a correction gain calculation means for calculating the correction gain and the correction gain calculation means is further provided, and the target value calculation means is provided by the correction gain calculation means. Based on the correction gain calculated as above, each of the above-mentioned corrected target values is calculated by correcting the above sub-target value. You.
  • the correction gain is set so that all the normalized change amounts are the same based on the reference amount deviation and the deviation derivative value. Each is calculated.
  • a modified sub-target value is calculated from the corrected target value.
  • a control system that further generalizes the first invention and the second invention provides a plurality of actuators for driving a control target. Corresponding to multiple actuary nights And a plurality of sensors for measuring the amount of control of the object to be controlled and the amount of control measured by the plurality of sensors. Calculate the manipulated variables so as to approach the given target value, and calculate these manipulated variables in the above-mentioned multiple factories overnight.
  • a control system equipped with a controller to provide the control amount the control amount measured by the plurality of sensors described above and the control amount Based on the final target values, the target values are calculated so that all the control amounts respond at the same rate, and these target values are calculated. It is characterized in that it is equipped with a target value generation device that gives values to the above controller.
  • a control method in which the first invention and the second invention are further generalized includes a plurality of actuators for driving the control target, and a plurality of the above-described plurality of actuators.
  • a plurality of sensors and a plurality of sensors are provided corresponding to each of the factories and measure the control amount of the above-mentioned controlled object. Then, the control amounts measured by are calculated so as to approach the given target value, and the control amounts are calculated as above. Set up a controller to give each of the number of actuaries overnight.
  • measurements are made by multiple sensors.
  • the target values are calculated based on the control amounts, so that all the control amounts respond to the same final target value at the same ratio as the given value. These target values are given to the controller.
  • the target value is calculated so that the rate of change of the control amount, that is, the response speed of the control amount, is the same, the response is fast.
  • the response is adjusted so that it is slow, and the response is slow so that it is fast. As a result, the response speed becomes uniform, so that the control target can be controlled stably.
  • Fig. 1 is a block diagram showing the overall configuration of the temperature control system of the molding machine.
  • Fig. 2 shows the arrangement of multiple heaters that make up a plate heater and the temperature sensors that can be installed for each heater. It is a figure.
  • Fig. 3 shows a graph of the measured temperature when the target temperature given to the multipoint temperature control device is constant.
  • Fig. 4 is a functional block diagram showing the detailed configuration of the target temperature generator.
  • Fig. 5a shows the fuzzy rule of the corrected gain inference knowledge
  • Fig. 5b shows the antecedent variable “reference value deviation” of the fuzzy rule.
  • the five member-ship functions representing the linguistic information “NB”, “NS”, “ZR”, “P5” and “jun8” are shown.
  • Figure 5c shows the antecedent variable “Fuzzy fine” of the fuzzy rule.
  • the five main functions that represent the linguistic information "NB”, “NS”, “ZRj,””PS” and “PB” for the "minute” are shown.
  • d shows the linguistic information "NB”, “NMj,”"NSj,””ZE”,”PS", and "L” for the consequent variable "correction gain” of the fuzzy rule.
  • Two singletons that represent “PMj and“ PB ” are shown.
  • Fig. 6a shows the fuzzy rule of settling inference knowledge
  • Fig. 6b shows the relation between the antecedent variable "normalized deviation” of the fuzzy rule
  • Fig. 6c shows the five main functions which represent the linguistic information "NS”, “ZR”, “PS”, “PM” and “PB”. Shows the linguistic information "NS”, “ZR”, “PS”, “PM” and “PB” for the consequent variable "settling degree” of the fuzzy rule. Shows five singletons.
  • Fig. 7 is a flow chart showing a series of processing steps for target temperature generation in the target temperature generator.
  • Fig. 8a is a graph showing the target temperature generated by the target temperature generator
  • Fig. 8b is a graph showing the target temperature shown in Fig. 8a at multiple points. This is a graph that indicates the measured temperature when the control device operates.
  • Figure 9 is a diagram showing the overall configuration of the position control system of the Akai Crane.
  • Fig. 10a is a graph showing the command speed generated by the conventional control
  • Fig. 10b is a graph showing that the command speed was controlled by the command speed shown in Fig. 10a
  • Fig. 10c is a graph showing the position of the hook at the moment, and is controlled by the command speed shown in Fig. 10a. This is a graph that shows the trajectory of the trolley and hook when it is set.
  • Fig. 11 is a function block diagram showing the detailed configuration of the command speed generator.
  • FIG. 12 is a flow chart showing a procedure for generating a command speed in the command speed generating device.
  • Fig. 13a is a graph showing the command speed generated by the command speed generator
  • Fig. 13b is a graph that is controlled by the command speed shown in Fig. 13a
  • Fig. 13c is a graph showing the position of the hook at the time of control. This is a graph showing the trajectory of the work.
  • Fig. 14a shows that when the maximum speed of the command speed is changed according to the travel distance in the conventional control and the drive wheel of the main girder slips
  • Fig. 14b shows the trolley when the main girder and the trolley move at the moving speed shown in Fig. 14a. And a graph showing the trajectory of the hook.
  • Fig. 15a shows the command speed generated by the command speed generator when the driving wheel of the main girder slips, similar to the graph shown in Fig. 14a. Is a graph showing the moving speed of the main girder and the trolley, and Fig. 15b shows the trolley when the trolley moves at the moving speed shown in Fig. 15a. This is a graph showing the trajectory of the hook and hook. Best form to carry out the invention
  • Fig. 1 is a block diagram showing the overall configuration of the temperature control system of the molding machine.
  • the upper plate H arranged above and below the conveyance path H: and the lower plate H It is heated by 2. Thereafter, the heated work W is press-formed by the upper die Pi and the lower die P2 of the press machine.
  • Each of the plate heaters H1 and H2, which heats the heater W consists of a plurality of heaters, as shown in Fig. 2.
  • Plate heater H1 is connected to heaters H1, H1, H2, H1,
  • Heater H 1 1, ⁇ 1, 2, ⁇ 1, 3, ⁇ 1,, ⁇ 1, 5 ⁇ ⁇ 1, 6, ⁇ 1,
  • the ambient temperature of the transfer path (approximately equal to the temperature of the work W) should be set at the space corresponding to the center of each night. Temperature sensors to measure the temperature of each of them. That is, H !, Hi, 2 ,
  • H, 3, H1,4, H1,5, H1,6, H! , 7 and corresponding temperature sensors S 1,! , S1,2, S1,3, S1,4, S1,5, S1.6, S1,7 and S!, 8 are provided and these temperature sensors are provided.
  • the temperature is measured at a fixed distance upward from the work W by a distance.
  • heater H2 ,! , H 2, 2, H 2, 3, H 2, 4, h 2, 5, H 2, 6, H 2, 7 and H 2, 8 correspond to the sensors S 2 , 1, S 2, 2, S 2, 3, S 2, 4, S 2, 5, S 2. 6, S 2. 7 your good beauty S 2. 8 is set only et al. are, in temperature was down service Therefore, the temperature at a position that is a fixed distance below the work W is measured.
  • thermocouples are, for example, thermocouples.
  • 5 2, 7, and S 2, 8 Measured temperature measured by each of ⁇ 1.1, ⁇ 1, 2, ⁇ 1, 3, ⁇ 1.4. , ⁇ ⁇ ! , 5, ⁇ ⁇ 1, 6, ⁇ ⁇ 1, 7, ⁇ ⁇ ! , 8, ⁇ ⁇ 2.1, ⁇ ⁇ 2, 2, ⁇ ⁇ 2, 3, ⁇ ⁇ 2, 4, ⁇ ⁇ 2, 5, ⁇ ⁇ 2, 6, ⁇ ⁇ 2, 7 and ⁇ ⁇ 2 . 8 is provided to the multipoint temperature control device 10. In addition, the measured temperature is given from the multipoint temperature control device 10 to the target temperature generation device 20 ⁇ ).
  • the multipoint temperature control device 10 includes a target temperature generation device described later.
  • the manipulated variables U i, j are determined so that the measured temperatures MT i, j become the target temperature TT i. Provided by the target temperature generator 20.
  • the multipoint temperature control device 10 is, for example, a discrete time type control device that performs temperature control by PID control.
  • the control cycle of the multipoint temperature control device 10 is, for example, 2 [sec].
  • Measuring temperature MT of heaters H 1 and H 5 located at the center compared to T 1 and 6! , 5 is fast. O one server one is measuring temperature MT 5 of Heater H 5 you position in central in was or sheet-menu door also large Good. This means that the heels ,, 5 are the nights around it, H!, 3, H1,4, H1,5, H! , And 7 from heat (interference).
  • the measured temperature MT i of i, j is generated so that the response of “,” is a fixed percentage.
  • a reference target temperature serving as a reference of each target temperature given to the multipoint temperature control device 10 is input. That is, Hey Evening Hi, H !, 2 , H !, 3 , H
  • T2.8 is input to the target temperature generator 20.
  • the start-up correction gain K 0 and the set-up correction gain are also input to the target temperature generation device 20.
  • Start-up correction gain K The purpose of this is to adjust the gain of the reference target temperature, which corrects the target temperature at the time of startup (transient state), and to adjust the gain according to the control target.
  • the optimal value is set.
  • the correction gain for settling is used to adjust the reference target temperature correction gain that corrects the target temperature during settling (steady state). Therefore, the optimal value is set according to the control target.
  • the start-up correction gain is 0 to 1, and the default value when the start-up correction gain is not input is 1.
  • Fig. 4 shows the detailed configuration of the target temperature generator 20
  • the target temperature generator 20 performs the reference value deviation calculation processing 21, the differential differentiation calculation processing 22, the normalized deviation calculation processing 23, the fuzzy inference processing 24, and the correction quantity calculation processing. 25 and target temperature calculation and processing 26.
  • the target temperature generator 20 is a computer -Implemented by a system and software running on it. Each process is a program routine
  • Generating periodic T S et is Ru is set Tsu by the control to scan te-time depending optimal generate periodic Gao Bae-les-over data.
  • the generation period Tset is 2 to 60 [sec], for example, 20 [sec].
  • the reference value deviation £ i.j, k is the deviation between the normalized measurement temperature NTi, j, k and the reference value NTBi, k .
  • the reference target temperature BT i, may be set to a different value for each of Hi, j, and k , the measured temperature MT i,”
  • k is the reference target temperature. It is normalized by the temperature BT i, j.
  • the normalized measurement temperature NT j, k indicates how much the measurement temperature MT ij k of each heater H has reached the reference target temperature BT ij x.
  • Reference value NTB i. K is Ru is determined roup-per (flop les over preparative Heater H 1 or the H 2).
  • Reference value NTB i. K is, is that Do different determine how stand on when or settling Ru Oh at (transient state) and One by the one Ru Oh in (steady state).
  • the reference values NTB i, k may be determined at the time of settling or by the same method as at startup.
  • the start-up time may be other than the set-up time.
  • the reference value NTB k is the normalized measurement temperature NT i. j, k may be the lowest value.
  • FIGS. 5a to 5d show an example of the correction gain inference knowledge set in advance in the fuzzy inference processing 24.
  • FIG. In this modified gain inference knowledge the subscripts i, j, and k of the antecedent and consequent variables are omitted.
  • Fig. 5a is an example of the fuzzy rule of the corrected gain inference knowledge.
  • Figure 5b shows the linguistic information “ ⁇ (Negative Big)” on the antecedent variable “reference value deviation ⁇ ” of the fuzzy rule of corrected gain inference knowledge.
  • Factory NS Negative Smal l
  • Factory ZR Zero
  • PS Prositive Smal l
  • PB Personal Big
  • Figure 5c shows the linguistic information “ ⁇ ”, “NS”, and “ ⁇ ” for the antecedent variable “differential difference d ⁇ ” of the fuzzy rule of the corrected gain inference knowledge. This is an example of the five main-ship functions representing “ZR”, “jun3” and “jun8”.
  • Figure 5d shows the linguistic information "NB" and the factory NM (Negative N.N.) Medium) j, factory NS, ZR, factory PS, PM (Positive Medium) and PB are examples of seven singletons.
  • the correction gain i.j, k increases the deviation ⁇ i, ”, k between the normalized measurement temperature NTi, j, k and the reference value ⁇ ⁇ i, j, k Nearly as rather come large.
  • normalized measured temperatures NT i, j. K is a reference value NTB i, j, deviation between the k £ i, j, k is that Do small Ku etc. ho small Ku
  • FIGS. 6a to 6c show an example of the settling degree inference knowledge set in advance in the fuzzy inference processing 24.
  • FIG. In this settling inference knowledge, the subscripts i, j, and k of the antecedent and consequent variables are omitted.
  • Figure 6a is an example of a fuzzy rule of settling degree inference knowledge.
  • Figure 6b shows the linguistic information “NS”, “ZR”, and “PSR” for the precondition variable “normalized deviation e” of the fuzzy rule of settling inference knowledge. ”,“ PM ”, and“ PB ”are examples of the five main functions.
  • Figure 6c shows the linguistic information “NS”, “ZR”, and “PS” for the consequent variable “settling /?” Of the fuzzy rule of settling inference knowledge.
  • ”,“ PM ”and“ PB ” are examples of the five singletons.
  • the absolute values of the setting degrees i, j, and k are closer to 0, the measured temperatures MT i, j, and k are set to the reference target temperatures BT i and j (at the time of setting; steady state ) .
  • the correction amounts Gi, j, and k are calculated based on the correction gain for setting and the correction gain.
  • et al given gills are Ru compensation amount G i. j. use have eyes Ri by the and the child you modify the Atsushi Shimegi TT i to 3 ⁇ 4, j. k in a, respectively Re its to be you de San is there.
  • the target temperature is generated by the target temperature generation device 20.
  • FIG. 7 is a flow chart showing a series of processes of target temperature generation in the target temperature generation device 20.
  • the generation of the target temperature this time is defined as the processing cycle k.
  • FIG. 2 is a diagram showing an overall configuration of a position control system of the present invention.
  • Two parallel rails RL running in the direction are those rails that are located at appropriate heights.
  • the main girder MB is supported by RL on the rail RL.
  • Rails for main girder M B are those rails that are located at appropriate heights.
  • This rail RL y is X Direction and the straight line to that person toward the door to the Rere Ru of rail RL y Te beauty of (the Re This will have a Y Direction) to Russia Li TR Les over RL y is supported by the move itself.
  • Rolls for TR The upper machine is fixed.
  • a hook FK for attaching a suspended load is attached to the tip of the roller W to be raised and lowered by the winding machine.
  • the main girder MB is Ru is going dynamic run in Tsu by the Bei gills the run line for the motors M x in the main girder MB. Ru is detect in Tsu by the main girder MB of position X of the eye that you detect the number of rotation row for the motors M run error down co-over da E x.
  • the position X of the main digit MB detected by the encoder E (hereinafter referred to as measurement position X) is given to the motor control device 41, It is also given to the command speed generator 50.
  • the trolley TR is a traversing motor provided in the trolley TR.
  • a measurement position Y (hereinafter, referred to as a measurement position Y) is given to a motor control device 41 and further to a command speed generator 50.
  • the motor control device 41 uses the command speed generation device 50 to change the moving speed MV of the main girder MB based on the measurement position X detected by the encoder E.
  • the driving mode M is controlled so that it becomes equal to the command speed SP given from the vehicle.
  • mode one evening control equipment 42 moving speed MV y of error down co-over da E y rather than based on Dzu to'm
  • One is detected measurement position location Y to collected by filtration Li TR is,
  • Directive The traverse motor My is controlled so that it becomes equal to the command speed SPy given from the speed generator 50.
  • the moving speed may be detected by setting a speed sensor at each motor and evening.
  • the hoist is equipped with a lifting / lowering motor (not shown) for raising and lowering the wire W.
  • a lifting / lowering motor (not shown) for raising and lowering the wire W.
  • the lifting / lowering motor rotates, and the hook FK is raised / lowered.
  • the command speed generation device 50 starts operating from a position (hereinafter referred to as an initial position (X., Y .;)) before carrying by the ceiling crane.
  • a position hereinafter referred to as an initial position (X., Y .;)
  • Ru base Ki finger Ordinance speed (SP X, SP y) is also of the Ru Oh that generates a.
  • Fig. 10a the command speed is independently generated in each direction from the initial position to the target position according to the preset speed pattern.
  • Fig. 10b is a graph showing the command speed according to the conventional control, and Fig. 10b shows the hook when the command speed is controlled by the command speed shown in Fig. 10a.
  • This is a graph showing the position of FK
  • Fig. 10c is a graph showing the position of the trolley TR (measured position) and the position of the box FK. .
  • the speed of the command speed under this conventional control is the first speed.
  • the initial position of the collected by filtration Li TR in the first 1 (X., Y.) Or Jo Luo operating speed (maximum speed) V ss to Do that or in full click FK is Speed up as much as possible to avoid swinging.
  • the speed is maintained.
  • decelerate and stop so that the hook FK does not swing and stop at the eye position (Xf, Yf ).
  • the distance or the time required from the main digit MB and the trolley TR force s to the maximum speed V ss to stop (speed 0) is previously determined. Therefore, the timing (position) at which to start deceleration is determined based on the target position.
  • the moving direction (velocity vector) of the hook FK changes abruptly, and the change in the moving direction leads to a change in the state of the coil. Acting on the hook FK, the hook FK swings vertically in the X-axis direction.
  • the hook FK vibrates in this way, even if the trolley TR reaches the target position, the suspended load is lowered until the vibration of the hook FK stops. Cannot be removed and the transfer time is too long. You. In addition, because the hook FK oscillates, it is necessary to extend the safety area beyond the working area.
  • Directive velocity generating equipment 50, et emissions co over da E x, Re Resona is their in One by the E y detection metrology position location X, Y (hereinafter, the measuring position location (X, Y) in and have groups Dzu Table to), the control panel or other input equipment ( Figure shown shown) or al entered by Ru at Nozomu of targets position location (X f of that, the Y f) And command speeds (SP x , SP y ) for transporting the trolley TR so that the suspended load on it can be moved as little as possible. That is what it is.
  • the command speed generator 50 generates a command speed at a fixed generation period T set. This processing cycle is represented by k.
  • FIG. 11 is a functional block diagram showing a detailed configuration of the command speed generating device 50.
  • the command speed generating device 50 is realized by the combo-evening and the software running on the same, and the entirety of the command-velocity generating device 50 is controlled by the noise window. It can also be realized.
  • the command speed generation device 50 can also be realized in part by software and in other parts by a noise window. Wear.
  • the command speed generation device 50 includes a travel distance calculation process 51, a normalization process distance calculation process 52, a reference value deviation calculation process 53, a deviation differentiation calculation process 54, and a fuzzy inference process. 55, correction amount calculation processing 56, speed pattern generation processing 57, and command speed calculation processing 58 are provided. Each process is a program routine.
  • the desired eye position (XY f ) is input by the operator to the moving distance calculation processing 51, the measurement positions from the encoders E x and E y are obtained.
  • the position (X, Y) is taken as the initial position (X, Y.) of the trolley TR, and is transferred to the moving distance calculation processing 51 via the motor control devices 41 and 42. It is.
  • the maximum travel distances for the X and ⁇ axis directions are calculated by the travel distance calculation processing 51. It is done.
  • the maximum travel distance (Lx, Ly) is calculated by the following equation.
  • Goal position location control is initiated to Chi was being entered, the mobile measurement position (X, Y) at the measurement position (X k, Y k) and between to generate periodic T se t septum Incorporated into distance calculation processing 52.
  • the distance from the initial position (X 0, Y.) based on the acquired measurement position (X k , Y k) is given from the movement distance calculation processing 51.
  • the maximum moving distance (L, L y) has been by Ri normalized
  • normalized distance (NL k, NL y. k ) is Ru are de San Ri by the normalized distance calculated out processing 52.
  • Normalized distance (NL, k, NL y, k) is Ru are de San Ri by the following equation.
  • Normalized distance (NL, NL y) of this is, whether collected by filtration Li TR is you are reachable throat of the extent to goal position location table to (goal achievement rate, moving distance in each square direction Even if the distance is different, a regular rule that can express 0 to 1 whether the trolley TR is close to or far from the target position based on the normalized distance can be expressed. The closer the activation distance is to 1, the closer to the eye position
  • Reference value NLB k is Ru most DaiUtsuri moving distance (L x, L y) the value of either Re not have the next depending on the of the atmosphere of the capital o
  • NLB k NLy, k ... (14)
  • This reference value is, in short, as shown in Fig. 10C, when the large movement distance Lx> Ly, the Y-axis direction is X Since the target position is reached faster than in the axial direction, the maximum speed is set in the X-axis direction. In order to reach the target at the same time as the X-axis direction, the Y-axis direction is moved at a speed smaller than the maximum speed. Is determined as follows.
  • the reference value may be determined in consideration of the maximum speed in addition to the maximum travel distance. Good.
  • Reference value deviation (£ x, k, £ y , k) is the reference value NLB k and the normalized distance (NL k, NL y, k ) as the in group Dzu les, Te de San Ri by the following equation It is.
  • the reference value deviation ( ⁇ k , ⁇ y , k ) calculated by the reference value deviation calculation processing 53 is given to a deviation differentiation processing 54 and a fuzzy inference processing 55. You.
  • the differential differentiation (d £ x , k , d ⁇ y , k ) is calculated by the differential differentiation process 54.
  • the deviation differential (d £ x , k, d £ y, k ) is calculated by the following equation. ⁇
  • the differential derivative (dx, k , d £ y , k ) time-differentiated by the differential differentiation process 54 is given to the fuzzy inference process 55.
  • the reference value deviation ( ⁇ > ⁇ , k ) and the deviation derivative (d, d ⁇ y are set in advance.
  • the corrected gain ( ay , k ) is calculated according to the corrected gain inference knowledge.
  • the corrected gain inference knowledge is shown in Figs. 5a to 5d. The same thing as knowledge is used.
  • the correction gain (a) calculated by the fuzzy inference processing 55 (also given to the correction amount calculation processing 56,
  • the correction amount calculation processing 56 the maximum movement distance (L, Ly ) given from the movement distance calculation processing 51 and the fuzzy inference processing 55 are used. Based on the correction gain (h, a) issued, the correction amount (G, G) of the command speed is calculated by the following equation.
  • Motor control equipment 41, 42 e Ru base Ki speed path Turn-given to emissions VP k is generate Ri by the speed path Turn-down production formation process 57, example given in the directive speed calculation origin management 58 Yes.
  • This velocity notation is generated based on the target position ( Xf , Yf ) and the measurement position (X, Y), and as shown in Fig. 10a, The pattern is generated so that the hook FK does not swing during acceleration and deceleration, and runs at the maximum speed VSS during constant speed running.
  • SP y, k VP k ⁇ G y, k ... (24)
  • This is these finger-old speed (SP, SP k) is finger-old speed (SP x, SP y) and the finger-old speed generator 50 and Luo motor control device 41, 42 Re given e la that in> mode one evening control equipment 41, 42 command speed (SP x, SP y) to be had based on Dzu mode - evening M, its the M y Each is controlled.
  • FIG. 12 is a flow chart showing a series of processes of command speed generation in the command speed generation device 50.
  • the current generation of the target temperature is defined as the processing cycle k.
  • the position control open Hajimemae moving distance calculated out processing goal position Tsu by the 51 (X f, Y f) and M o evening control equipment 41, 42 and through the E down co over da E x,
  • the initial position (X, Yo) given from Ey is taken, and the maximum travel distance (Lx, Ly) is calculated.
  • the command speed generation device 50 determines the measurement positions (X k, Y k) given from the encoders E and E y via the motor control devices 41 and 42. Capture at a fixed generation cycle T set (Step
  • the normalized distance calculation processing 52 calculates the travel distance from the initial position (X, Y) based on the acquired measurement position ( Xk , Yk ). This move distance As a maximum moving distance (L x, L y) to be Tsu by normalized distance is normalized (NL x, NL y) you out of the Re its been Resona calculated (scan tape Tsu Step 62).
  • Reference value deviation calculation output processing 53 normalized distance (NL k, NL y, k) in that determine, respectively Re its reference value NLB k to have group Dzu (scan STEP 63).
  • the reference value deviation calculation processing 53 includes a normalized distance (NL k , N
  • Deviation differential calculation output processing 54 now times of the group. Quasi-value deviation (£ X. K, ⁇ y, k ) and the previous reference value deviation (£ X -i, ⁇ y, k-1, the deviation derivative (d £ x , k , d ⁇ y ,, k ) Is calculated respectively (step 65).
  • the fuzzy inference processing 55 uses the pre-set correction gain inference knowledge to calculate the reference value bias ⁇ (6k, ⁇ £ y, k) and the differential derivative. (D £ x .k , d £ y , k) and calculate the correction gains (hy x, k, hi y, k) respectively (Step ° 66
  • the correction amount calculating process 25 is based on the speed gain VP k generated by the speed pattern generating process 57 based on the correction gain (hi k, y, k). Then, the correction amounts ( Gx , k , Gy , k ) are determined (step 67).
  • Directive rate calculated out processing 58 the correction amount of its speed path evening over emissions VP k (G, G y. ) Ri by the and the child you modified have use a finger-old speed (SP k, SP y, k ) are calculated (step 68).
  • Directive speed production formation apparatus 20 Directive speed (SP k, SP y, u )
  • Re Su is Resona directive speed (SP x. K, SP y . K) and to motor one evening control equipment 41 , 42 (step 69) o
  • Mode one evening control equipment 41, 42 finger-old speed generating equipment 50 or al output is Ru finger Ordinance speed (SP x, SP y) intends line the control of the mode one evening to have use the.
  • FIG. 13 a diagram shows an example of a directive speed as generate Ri by the directive speed generating equipment 20 (SP x, SP y)
  • 13 b figure 13 a shows the positions of the trolley TR and the hook FK when controlled by the command speed shown in Fig. 13.
  • Fig. 13c shows the command shown in Fig. 13a. Shows the positions of the trolley TR and hook FK when controlled by speed.
  • Figures 13a to 13b correspond to the conventional controls shown in Figures 10a to 10b, respectively.
  • the P speed is determined by the command speed SP in the Y-axis direction against the command speed SP x in the X-axis direction, and the command speed SP according to the travel distance. Due to the smaller size,
  • the hook FK moves almost linearly up to X f, Y f ). In this way, since the main digit MB and the trolley TR arrive at the target position at the same time, the movement direction of the hook FK does not change. , And it will not generate crisp force. Therefore, the runout occurring in the hook FK is suppressed to a minimum. Since the swing of the hook FK is suppressed to a small level, the lifting of the suspended load can be performed without waiting for the swing of the hook FK (the suspended load) to decrease. Since the lowering can be done, the working time is shortened, and the swing of the suspended load is small, so there is no need to set up a safety area outside the working area. Become.
  • the 1 4 b showing a 1 4 a view your good beauty is, arrives we be in to its being Resona Re the same time the main girder MB and collected by filtration Li TR is the goal position location (XY f)
  • the graph shows the control result by the conventional control in which the maximum degree is changed according to the ring travel distance.
  • FIG. 14a shows the transfer of the main girder MB.
  • the moving speed MV y of the dynamic speed MV and collected by filtration Li TR indicates
  • the driving direction of the hook FK changes depending on the slip of the driving wheel of the main girder MB (X direction), and thus, the corioca works on the hook FK.
  • the swing is occurring in the hook. Arrives we are lagging position to the goal position location of Bok D Li TR Ri due to the child of the scan Clip (X 3, Y f) to have your (time T 3) move of the off-click FK Due to the change of the vector again, a larger swing is occurring in the hook FK.
  • Fig. 15a and Fig. 15b show the command speed generated by the command speed generator 50 when a slip occurs as in the conventional control described above.
  • the graph shows a more controlled graph.
  • Fig. 15a when a slip occurs on the drive wheel of the main girder MB and the moving speed MV in the X direction decreases, the size of the slip increases.
  • Flip and Y direction finger-old speed SP y of direction is Ru is small Ku change.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Mathematical Physics (AREA)
  • Fuzzy Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Feedback Control In General (AREA)

Abstract

Des températures mesurées par des capteurs de température correspondant à une série d'éléments de chauffage sont normalisées en fonction d'une température cible de référence, pour obtenir une valeur de référence. L'écart présenté par chacune des températures normalisées par rapport à la valeur de référence est calculé et différencié dans le temps. Un gain de correction est calculé sur la base des écarts et des écarts différenciés. Les écarts entre les températures mesurées par rapport aux températures cibles de référence correspondantes sont normalisés en fonction de la température cible de référence. Le degré de stabilisation est calculé sur la base de ces trois écarts normalisés. Les valeurs de correction sont calculées sur la base des gains correspondants et du degré de stabilisation correspondant. Les températures cibles sont calculées en corrigeant la température cible de référence en fonction des valeurs de correction.
PCT/JP1995/000612 1994-03-31 1995-03-30 Systeme et methode de regulation WO1995027235A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP95913388A EP0753802A4 (fr) 1994-03-31 1995-03-30 Systeme et methode de regulation
KR1019960705531A KR970702515A (ko) 1994-03-31 1995-03-30 제어 시스템 및 방법(Control system and method)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP8379494 1994-03-31
JP6/83794 1994-03-31

Publications (1)

Publication Number Publication Date
WO1995027235A1 true WO1995027235A1 (fr) 1995-10-12

Family

ID=13812562

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP1995/000612 WO1995027235A1 (fr) 1994-03-31 1995-03-30 Systeme et methode de regulation

Country Status (4)

Country Link
EP (1) EP0753802A4 (fr)
KR (1) KR970702515A (fr)
CN (1) CN1144566A (fr)
WO (1) WO1995027235A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1998024009A1 (fr) * 1996-11-28 1998-06-04 Siemens Aktiengesellschaft Procede pour parametrer un automate a logique floue qui compare un signal de mesure a un signal modele
JP2003510676A (ja) * 1999-09-23 2003-03-18 ケイアイシー サーマル プロファイリング インコーポレイテッド コンベヤ式熱プロセッサ内の部品の温度レスポンスを制御する方法及び装置

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2004038776A1 (fr) * 2002-10-25 2004-05-06 Tokyo Electron Limited Systeme de traitement thermique et procede de traitement thermique
DE102005027435B4 (de) * 2005-06-14 2007-04-26 Siemens Ag Regelverfahren für eine Anzahl von in einem Regeltakt lagegeregelten Folgeachsen
CN100517135C (zh) * 2005-11-18 2009-07-22 鸿富锦精密工业（深圳）有限公司 自动控制模拟系统及自动控制模拟方法
JP5578990B2 (ja) * 2010-08-27 2014-08-27 アズビル株式会社 エネルギー総和抑制制御装置、電力総和抑制制御装置および方法
JP5641960B2 (ja) * 2011-02-01 2014-12-17 三菱電機株式会社 内燃機関の制御装置
US11199822B2 (en) * 2017-12-15 2021-12-14 Omron Corporation Control device
CN110147042B (zh) * 2019-05-28 2020-06-16 金力 一种基于模糊控制结合pid控制的直立agv车体控制方法
CN110727292B (zh) * 2019-10-10 2020-11-06 北京北特圣迪科技发展有限公司 一种舞台三维动态环系统运动的控制方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS58149512A (ja) * 1982-02-27 1983-09-05 Fanuc Ltd 数値制御方式
JPS61264407A (ja) * 1985-05-20 1986-11-22 Matsushita Electric Ind Co Ltd 角加速度制御方法
JPH03204008A (ja) * 1989-12-31 1991-09-05 S G:Kk 複数軸の同調制御方式
JPH04227502A (ja) * 1990-06-26 1992-08-17 Sekisui Chem Co Ltd ファジィ推論による温度制御システム
JPH04267401A (ja) * 1991-02-22 1992-09-24 Yamatake Honeywell Co Ltd 制御装置
JPH0596591A (ja) * 1991-10-11 1993-04-20 Sumitomo Heavy Ind Ltd 射出成形機の制御装置
JPH0679730A (ja) * 1992-08-31 1994-03-22 Nok Corp 加硫機の温度制御装置

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0462815B1 (fr) * 1990-06-21 1996-09-25 Honeywell Inc. Commande adaptative à horizon variable avec des moyens pour minimiser les coûts de fonctionnement
US5384698A (en) * 1992-08-31 1995-01-24 Honeywell Inc. Structured multiple-input multiple-output rate-optimal controller

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS58149512A (ja) * 1982-02-27 1983-09-05 Fanuc Ltd 数値制御方式
JPS61264407A (ja) * 1985-05-20 1986-11-22 Matsushita Electric Ind Co Ltd 角加速度制御方法
JPH03204008A (ja) * 1989-12-31 1991-09-05 S G:Kk 複数軸の同調制御方式
JPH04227502A (ja) * 1990-06-26 1992-08-17 Sekisui Chem Co Ltd ファジィ推論による温度制御システム
JPH04267401A (ja) * 1991-02-22 1992-09-24 Yamatake Honeywell Co Ltd 制御装置
JPH0596591A (ja) * 1991-10-11 1993-04-20 Sumitomo Heavy Ind Ltd 射出成形機の制御装置
JPH0679730A (ja) * 1992-08-31 1994-03-22 Nok Corp 加硫機の温度制御装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP0753802A4 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1998024009A1 (fr) * 1996-11-28 1998-06-04 Siemens Aktiengesellschaft Procede pour parametrer un automate a logique floue qui compare un signal de mesure a un signal modele
JP2003510676A (ja) * 1999-09-23 2003-03-18 ケイアイシー サーマル プロファイリング インコーポレイテッド コンベヤ式熱プロセッサ内の部品の温度レスポンスを制御する方法及び装置

Also Published As

Publication number Publication date
CN1144566A (zh) 1997-03-05
KR970702515A (ko) 1997-05-13
EP0753802A4 (fr) 1997-05-07
EP0753802A1 (fr) 1997-01-15

Similar Documents

Publication Publication Date Title
WO1995027235A1 (fr) Systeme et methode de regulation
US9383745B2 (en) Method and device for controlling furnace temperature of burning heating furnace
CN101973489B (zh) 带有图像传感器的双起升桥吊控制系统及控制方法
US5216342A (en) Sliding mode control method with a feedforward compensation function
CN109911771A (zh) 变系数自抗扰控制器设计方法、及吊车自抗扰控制器
CN102009880A (zh) 电梯
CN116931508A (zh) 一种基于运动模型的多轴智能伺服驱控一体控制系统
KR100293185B1 (ko) 흔들림방지기능을구비한무인천정크레인제어장치
CN102079481A (zh) 双起升双吊具场桥控制系统及其控制方法
EP2896590A1 (fr) Procede et dispositif pour contrôler en boucle ouverte l'oscillation de la charge pour engin rotatif de levage
KR20140000333A (ko) 엘리베이터의 도어 제어 장치
JPH1121077A (ja) クレーン制御装置
JP2002370418A (ja) プリント・ヘッド・モータの制御方法及びプリンタ
JP2861390B2 (ja) 制御システムおよび方法
CN109095328B (zh) 一种高速电梯轿厢水平振动的减振系统及其控制方法
US5219109A (en) Level control method for spliced strip materials having different unit weights
CN113023570A (zh) 悬挂式起重机的控制装置以及逆变器装置
JPH0475113A (ja) 制御装置
JP3265218B2 (ja) エレベータの制御装置
Toxqui et al. PD control of overhead crane systems with neural compensation
CN110817344A (zh) 一种输送积放链自整定系统
Rodriguez et al. Anti-swing control with hierarchical fuzzy CMAC compensation for an overhead crane
JP3285069B2 (ja) 張力制御方法
Laghrouche et al. Higher order sliding mode control based on optimal LQ control and integral sliding mode
CN214154374U (zh) 伺服电机及其驱动器、控制系统

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 95192292.0

Country of ref document: CN

AK Designated states

Kind code of ref document: A1

Designated state(s): CN JP KR US

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE CH DE DK ES FR GB GR IE IT LU MC NL PT SE

DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
121 Ep: the epo has been informed by wipo that ep was designated in this application
ENP Entry into the national phase

Ref document number: 1996 718308

Country of ref document: US

Date of ref document: 19960930

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 1995913388

Country of ref document: EP

WWP Wipo information: published in national office

Ref document number: 1995913388

Country of ref document: EP

WWW Wipo information: withdrawn in national office

Ref document number: 1995913388

Country of ref document: EP